From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263526AbTDCTjQ (for ); Thu, 3 Apr 2003 14:39:16 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263525AbTDCTjP (for ); Thu, 3 Apr 2003 14:39:15 -0500 Received: from ns1.scripps.edu ([192.42.82.59]:38328 "EHLO ns1.scripps.edu") by vger.kernel.org with ESMTP id S263524AbTDCTiQ (for ); Thu, 3 Apr 2003 14:38:16 -0500 Date: Thu, 3 Apr 2003 11:49:42 -0800 (PST) From: Andy Arvai Message-Id: <200304031949.LAA28555@astra.scripps.edu> To: linux-raid@vger.kernel.org Subject: Re: RAID 5 performance problems Cc: linux-kernel@vger.kernel.org X-Sun-Charset: US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Shouldn't /proc/mdstat have [UUUUU] instead of [_UUUU]? Perhaps this is running in degraded mode. Also, you have 'algorithm 0', whereas my raid5 has 'algorithm 2', which is the left-symmetric parity algorithm. Andy > cat /proc/mdstat gives: > > Personalities : [raid0] [raid5] > read_ahead 1024 sectors > md0 : active raid5 hdk1[4] hdi1[3] hdg1[2] hde1[1] hdc1[0] > 468872704 blocks level 5, 128k chunk, algorithm 0 [5/5] [_UUUU] > unused devices: